Lung cancer is a serious disease and causes many deaths around the world. Finding it early is very important because it helps in better treatment and improving patient survival rates. Doctors usually use methods like CT scans, biopsies, and manual checking, but these methods can be slow and may not always be accurate. They also depend a lot on the doctor’s experience and may miss cancer in the early stage. Artificial Intelligence (AI) helps improve diagnosis by automatically analyzing complex medical data. Deep Learning models can extract patterns from imaging and genomic data, thereby improving detection and classification performance. Bio-inspired algorithms, which are based on natural processes, identify the most relevant features and help enhance the model’s performance. Bio-inspired algorithms include Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO), and swarm behaviors to optimize feature selection. This research work presents a hybrid approach that combines AI-based models along with bio-inspired optimization techniques for lung cancer detection and classification. When AI and bio-inspired algorithms are used together, they yield better, faster results. This paper explains how these methods work together to detection and classification of lung cancer. The results show that combining these methods improves accuracy and helps doctors make better decisions.

This study proposed an approach that used deep learning techniques for feature extraction and bio-inspired methods, involving Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Particle Swarm Optimization (PSO), for optimal feature selection. The findings showed that the hybrid model significantly outperformed traditional methods, achieving higher accuracy and improved reliability. The use of multimodal data further enhanced early-stage detection and classification performance by combining imaging, clinical, and histopathological information. Overall, the study confirms that integrating AI and bio-inspired algorithms can serve as a powerful decision-support system, enhancing diagnostic accuracy and supporting the early detection of lung cancer.

[1] Qaraad, Mohammed, et al. “An optimized computational framework for non-small cell lung cancer subtype classification and biomarker discovery.” Artificial Intelligence Review (2025). [2] Tortora, Matteo, et al. “Deep reinforcement learning for fractionated radiotherapy in non-small cell lung carcinoma.” Artificial Intelligence in Medicine 119 (2021): 102137. [3] Roy, Avigyan, et al. “Adaptive genetic algorithm based deep feature selector for cancer detection in lung histopathological images: A. Roy et al.” Scientific Reports 15.1 (2025): 4803. [4] Karthika, M. S., Harikumar Rajaguru, and Ajin R. Nair. “Performance enhancement of classifiers through Bio inspired feature selection methods for early detection of lung cancer from microarray genes.” Heliyon 10.16 (2024). [5] Saleem, Muhammad Asim, et al. “Sooty tern optimization algorithm-based deep learning model for diagnosing NSCLC tumours.” Sensors 23.4 (2023): 2147. [6] Rao, Kokkila Gadda Koteswara, and Lalitha Kumari Pappala. “Early-Stage Multimodal Lung Cancer Detection Using Attention-Based Feature Fusion.” IEEE Access (2026). [7] Sandhiya, R., and R. Sasikumar. “Swallow Winged Kite Optimization with Shuffle Attention Xtreme Gradient Boost Network for lung cancer detection using CT image.” European Spine Journal (2025): 1-28. [8] Mohamed, Tehnan IA, et al. “A novel feature selection algorithm for identifying hub genes in lung cancer.” Scientific Reports 13.1 (2023): 21671. [9] Li, Kunpeng, et al. “Lung adenocarcinoma identification based on hybrid feature selections and attentional convolutional neural networks.” Mathem. Biosci. Eng 21 (2024): 2991-3015.